A research project is proposed to examine the interaction between nighttime sleep and daytime function (sleepiness/alertness and performance) in healthy elderly subjects. In addition to observing the relation between basal sleep parameters and daytime alertness across individuals, the project will test this relationship by manipulating nighttime sleep. The existence of a very sensitive measure of daytime sleepiness, the Multiple Sleep Latency Test, makes it possible to examine the effect of relatively small decreases or increases of nocturnal sleep, as well as the effect of total sleep loss. The overriding issue in this project is whether the well-described age-related changes in nighttime sleep are associated with parallel changes in daytime function--indeed, whether or not there is a causal relationship. Specifically, is the fragmentation and reduction of nocturnal sleep in at least some elderly persons part of a chronic sleep deprivation condition associated with non-optimal and potentially correctable levels of daytime alertness? Additional features of the project include: (1) Assessment of respiration, electrocardiogram, and leg movements during sleep with the goal of relating these parameters to sleep disturbance in the aged; and (2) 24-hour monitoring of body temperature and activity to evaluate underlying circadian rhythmicity.